Abstract Seizures are the most frequent overt manifestation of acute neurologic disorders in newborns, and seizures occur more commonly in newborns (2-3.5 per 1000 live births) than at any other age. In newborns, 75% of seizures are caused by an acute neurologic disorder such as hypoxic-ischemic encephalopathy (HIE). Both animal and human data show that seizures exacerbate neonatal hypoxic-ischemic brain injury; thus improved treatment of seizures may reduce the frequent and disabling long-term sequelae associated with neonatal seizures caused by HIE. Despite the high incidence and serious consequences of neonatal seizures, treatment is largely guided by empiric management, as only one randomized trial has been conducted of any antiepileptic drugs (AEDs) in newborns. That trial and other studies show that conventional AEDs are relatively ineffective in controlling refractory neonatal seizures. In addition to the lack of previous trials, there are important challenges to diagnosing and treating seizures quickly in critically ill newborns with HIE who have rapid onset of refractory seizures. To address these problems, the proposed trial will: 1) Test the feasibility of a novel trial design of early enrollment and rapid, continuous and prolonged EEG monitoring so that AEDs can be tested early in the course of neonatal seizures when more effective seizure control is most needed. 2) Evaluate the pharmacokinetics and safety of bumetanide (BTN) in newborns with refractory seizures caused by HIE in a Phase I Trial. Basic science research has shown BTN to be effective in reducing seizure activity in neonatal animals by blocking a specific chloride transporter which is highly expressed in the newborn brain of rodents and humans. Further experimental data show BTN to be particularly effective against seizures when used in combination with phenobarbital (PB), the standard first drug given to treat seizures in human newborns. In addition, BTN is a commercially available drug used safely in newborns as a diuretic for decades, and thus is an ideal novel AED to test in newborns. The study design will be a randomized, double-blind, controlled, dose escalation study of BTN as add-on therapy to treat refractory seizures caused by HIE not controlled by an initial loading dose of PB. The trial will test the feasibility of early enrollment of newborns with HIE, rapid application of a full montage EEG, and continuous review of EEG data to detect refractory seizures as soon as they occur following an initial loading dose of PB. When an EEG-proven seizure occurs at least 30 minutes following a loading dose of PB, the newborn will be randomized to receive either BTN or placebo in conjunction with a second dose of PB. Clinical, laboratory and continuous EEG monitoring data obtained after BTN administration will be analyzed to determine the pharmacokinetics and safety of BTN by comparing data from treatment and standard therapy groups. This study address important challenges in trial design and sets the stage for trials to improve treatment of neonatal seizures. Data from this pilot study will be used to guide design of a planned Phase III multicenter trial to test the efficacy of BTN to control refractory neonatal seizures.